Fingerprint biometric detection method, device, and storage medium

Abstract

The embodiments of the present application disclose a fingerprint biometric detection method, device, and storage medium, which fall within the field of security technology. In the embodiments of the present application, after performing region segmentation on a first fingerprint image and a second fingerprint image, a multi-partition grayscale distribution feature vector is determined based on the segmented regions, and fingerprint biometric detection can be performed through a fingerprint biometric detection model, wherein the first fingerprint image is a red channel image and the second fingerprint image is a blue channel image. Since the grayscale value distribution rule of each region of the two-channel images of a non-biometric fingerprint differs from that of a biometric fingerprint, it is possible to recognize whether the grayscale distribution feature of a biometric fingerprint exists in the fingerprint image through the multi-partition grayscale distribution feature vector, and furthermore, to determine whether the said biometric fingerprint is a biometric fingerprint, thereby providing higher security.

Description

Technology Field

[0001] The embodiments of the present application claim priority to a Chinese patent application filed on April 30, 2020, with application number 202010364403.X and title of invention "method, device and storage medium for fingerprint biometric detection", the entire contents of which are incorporated into the embodiments of the present application by reference.

[0002] The embodiments of the present application relate to the field of security technology, and in particular to a method, apparatus, and storage medium for fingerprint biometric recognition. Background Technology

[0003] Currently, fingerprint recognition is critically applied in many scenarios, such as attendance tracking, access control, computer unlocking, mobile phone unlocking, and payments. Fingerprint recognition primarily determines the validity of a fingerprint by extracting features from the image, such as patterns, breakpoints, and intersections. However, fingerprint forgery technologies are also continuously evolving; for instance, silicone fingerprint films are created to forge another person's fingerprint, allowing for successful bypasses of fingerprint recognition to commit illegal acts. To ensure the security of fingerprint recognition, it is necessary to incorporate biometric fingerprint detection capabilities based on the recognition process to effectively prevent the impersonation of another person's fingerprint using fingerprint films.

[0004] Previously, it was possible to detect whether a fingerprint was biometric based on skin temperature or skin capacitance value, but since skin temperature and skin capacitance value are generally relatively stable, malicious actors can easily forge fingerprint films with the same skin temperature or capacitance value to successfully bypass fingerprint biometric detection, so the security of these methods is still low.

[0005] Embodiments of the present application provide a fingerprint biometric detection method, a device, and a storage medium capable of improving the stability of a fingerprint biometric detection method. The technical solution is as follows:

[0006] In one embodiment, a fingerprint biometric detection method is provided, said method:

[0007] A step of acquiring a first fingerprint image and a second fingerprint image - the first fingerprint image and the second fingerprint image are images of the same fingerprint, the first fingerprint image is a red channel image, and the second fingerprint image is a blue channel image -;

[0008] A step of performing region segmentation for each of the first fingerprint image and the second fingerprint image - the segmented first fingerprint image includes a plurality of first predicted bright areas and a plurality of first predicted dark areas, and the segmented second fingerprint image includes a second predicted bright area and a plurality of second predicted dark areas -;

[0009] A step of determining a multi-partition gray level distribution feature vector based on the plurality of first predicted bright regions, the plurality of first predicted dark regions, the plurality of second predicted bright regions, and the plurality of second predicted dark regions;

[0010] The step of obtaining a fingerprint biometric detection result by inputting the above multi-partition grayscale distribution feature vector into a fingerprint biometric detection model - the fingerprint biometric detection model is obtained by training through supervised learning based on a biometric fingerprint sample image and a non-biometric fingerprint sample image -; includes.

[0011] Optionally, the step of acquiring the first fingerprint image and the second fingerprint image is,

[0012] Step of acquiring a first original image - the first original image is an image of a fingerprint collected with the green light turned off and the red and blue lights turned on -;

[0013] A step of generating the first fingerprint image by extracting image data of the red channel of the first original image;

[0014] The method includes the step of generating a second fingerprint image by extracting image data of the blue channel of the first original image.

[0015] Optionally, the step of acquiring the first fingerprint image and the second fingerprint image is,

[0016] Step of acquiring the second original image and the third original image - the second original image is a fingerprint image collected with the green and blue lights off and the red light on, and the third original image is a fingerprint image collected with the green and red lights off and the blue light on -;

[0017] A step of generating the first fingerprint image by extracting image data of the red channel of the second original image;

[0018] It includes the step of generating a second fingerprint image by extracting image data of the blue channel of the third original image.

[0019] Optionally, the step of performing region segmentation for each of the first fingerprint image and the second fingerprint image is,

[0020] A step of determining partition parameters based on the distribution rules of bright and dark regions of the red channel and blue channel images of the biometric fingerprint - the partition parameters include multiple partition width ratios -;

[0021] The method includes the step of performing region division for each of the first fingerprint image and the second fingerprint image based on a plurality of partition width ratios.

[0022] Optionally, the image sizes of the first fingerprint image and the second fingerprint image are the same, and the step of determining a multi-partition grayscale distribution feature vector based on the plurality of first predicted bright areas, the plurality of first predicted dark areas, the plurality of second predicted bright areas and the plurality of second predicted dark areas, is

[0023] A step of determining a first grayscale feature vector based on the plurality of first predicted bright regions and the plurality of second predicted dark regions;

[0024] A step of determining a second grayscale feature vector based on the plurality of first predicted dark regions and the plurality of second predicted bright regions;

[0025] It includes the step of generating the multi-partition grayscale distribution feature vector based on the first grayscale feature vector and the second grayscale feature vector.

[0026] Optionally, the step of determining a first grayscale feature vector based on the plurality of first predicted bright regions and the plurality of second predicted dark regions is,

[0027] A step of determining the number of pixel points corresponding to each grayscale value within the plurality of first predicted bright regions and the plurality of second predicted dark regions;

[0028] The method includes the step of generating the first grayscale feature vector based on the number of pixel points corresponding to each grayscale value within the plurality of first predicted bright regions and the plurality of second predicted dark regions.

[0029] Optionally, the step of determining a second grayscale feature vector based on the plurality of first predicted dark regions and the plurality of second predicted bright regions is,

[0030] A step of determining the number of pixel points corresponding to each grayscale value within the plurality of first predicted dark regions and the plurality of second predicted bright regions;

[0031] It includes the step of generating a second grayscale feature vector based on the number of pixel points corresponding to each grayscale value within a plurality of first predicted dark regions and a plurality of second predicted bright regions.

[0032] Optionally, before determining the first grayscale feature vector based on the plurality of first predicted bright regions and the plurality of second predicted dark regions,

[0033] The method further includes the step of obtaining a grayscale value statistical range - said grayscale value statistical range is determined based on the exposure parameters of an image collection device that collects the collected fingerprint image -;

[0034] The step of determining a first grayscale feature vector based on the plurality of first predicted bright regions and the plurality of second predicted dark regions is,

[0035] A step of determining the number of pixel points corresponding to each grayscale value within the grayscale value statistical range in the plurality of first predicted bright regions and the plurality of second predicted dark regions based on the above grayscale value statistical range;

[0036] The method includes the step of generating the first grayscale feature vector based on the number of pixel points corresponding to each grayscale value within the grayscale value statistical range in the plurality of first predicted bright regions and the plurality of second predicted dark regions.

[0037] Optionally, the step of generating the multi-partition grayscale distribution feature vector based on the first grayscale feature vector and the second grayscale feature vector is,

[0038] It includes the step of obtaining a multi-partition grayscale distribution feature vector by joining the first grayscale feature vector and the second grayscale feature vector.

[0039] Optionally, prior to the step of inputting the multi-partition grayscale distribution feature vector into a fingerprint biometric detection model to obtain a fingerprint biometric detection result,

[0040] A step of acquiring the above-mentioned biometric fingerprint sample image and the above-mentioned non-biometric fingerprint sample image - the above-mentioned biometric fingerprint sample image includes images of a plurality of biometric fingerprint samples collected with the green light turned off and the red light and blue light turned on, and the above-mentioned non-biometric fingerprint sample image includes images of a plurality of non-biometric fingerprint samples collected with the green light turned off and the red light and blue light turned on -;

[0041] It further includes the step of obtaining the fingerprint biometric detection model through supervised learning based on the above-mentioned biometric fingerprint sample image and non-biometric fingerprint sample image.

[0042] Optionally, the above method is,

[0043] Step of acquiring the fourth original image - the fourth original image is an image of the fingerprint collected while the green light is on -;

[0044] A step of determining a fingerprint recognition result based on the above-mentioned fourth original image;

[0045] It further includes a step of determining a security verification result based on the fingerprint recognition result and the fingerprint biometric detection result.

[0046] In another aspect, a fingerprint biometric detection device is provided, wherein the device,

[0047] A first acquisition module for acquiring a first fingerprint image and a second fingerprint image - the first fingerprint image and the second fingerprint image are images of the same fingerprint, the first fingerprint image is a red channel image, and the second fingerprint image is a blue channel image -;

[0048] A partition module for performing region division for each of the first fingerprint image and the second fingerprint image - the divided first fingerprint image includes a plurality of first predicted bright areas and a plurality of first predicted dark areas, and the divided second fingerprint image includes a second predicted bright area and a plurality of second predicted dark areas -;

[0049] A first determination module that determines a multi-partition grayscale distribution feature vector based on the plurality of first predicted bright regions, the plurality of first predicted dark regions, the plurality of second predicted bright regions, and the plurality of second predicted dark regions;

[0050] A detection module that inputs the above multi-partition grayscale distribution feature vector into a fingerprint biometric detection model to obtain a fingerprint biometric detection result - the above fingerprint biometric detection model is obtained by training through supervised learning based on a biometric fingerprint sample image and a non-biometric fingerprint sample image -; is included.

[0051] Optionally, the first acquisition module is,

[0052] A first acquisition module for acquiring a first original image - the first original image is an image of a fingerprint collected with the green light off and the red and blue lights on -;

[0053] A first extraction unit that generates the first fingerprint image by extracting image data of the red channel of the first original image;

[0054] It includes a second extraction unit that extracts image data of the blue channel of the first original image and generates a second fingerprint image.

[0055] Optionally, the first acquisition module is,

[0056] A second acquisition unit acquiring a second original image and a third original image - the second original image is a fingerprint image collected with the green and blue lights off and the red light on, and the third original image is a fingerprint image collected with the green and red lights off and the blue light on -;

[0057] A third extraction unit that generates the first fingerprint image by extracting image data of the red channel of the second original image;

[0058] It includes a fourth extraction unit that extracts image data of the blue channel of the third original image to generate a second fingerprint image.

[0059] Optionally, the above partition module,

[0060] A first determination unit that determines partition parameters based on the distribution rules of bright and dark regions of the red channel and blue channel images of a biometric fingerprint—partition parameters include multiple partition width ratios—;

[0061] It includes a partition unit that performs area division for each of the first fingerprint image and the second fingerprint image based on a plurality of partition width ratios.

[0062] Optionally, the image sizes of the first fingerprint image and the second fingerprint image are the same, and the first determination module,

[0063] A second determination unit for determining a first grayscale feature vector based on the plurality of first predicted bright regions and the plurality of second predicted dark regions;

[0064] A third determination unit that determines a second grayscale feature vector based on the plurality of first predicted dark regions and the plurality of second predicted bright regions;

[0065] It includes a generating unit that generates the multi-partition grayscale distribution feature vector based on the first grayscale feature vector and the second grayscale feature vector.

[0066] Optionally, the second decision unit is,

[0067] A first determination sub-unit for determining the number of pixel points corresponding to each grayscale value within the plurality of first predicted bright areas and the plurality of second predicted dark areas;

[0068] It includes a first generating sub-unit that generates the first grayscale feature vector based on the number of pixel points corresponding to each grayscale value within the plurality of first predicted bright regions and the plurality of second predicted dark regions.

[0069] Optionally, the third decision unit,

[0070] A second determination sub-unit for determining the number of pixel points corresponding to each grayscale value within the plurality of first predicted dark regions and the plurality of second predicted bright regions;

[0071] It includes a second generating sub-unit that generates a second grayscale feature vector based on the number of pixel points corresponding to each grayscale value within the plurality of first predicted dark regions and the plurality of second predicted bright regions.

[0072] Optionally, the first decision module is,

[0073] A third acquisition unit for acquiring a grayscale value statistical range—the grayscale value statistical range is determined based on the exposure parameters of an image acquisition device that collects the collected image of the fingerprint—; further comprising;

[0074] The above second decision unit is,

[0075] A third determination server unit that determines the number of pixel points corresponding to each grayscale value within the grayscale value statistical range in the plurality of first predicted bright areas and the plurality of second predicted dark areas based on the grayscale value statistical range;

[0076] It includes a third generating sub-unit that generates the first grayscale feature vector based on the number of pixel points corresponding to each grayscale value within the grayscale value statistical range in the plurality of first predicted bright regions and the plurality of second predicted dark regions.

[0077] Optionally, the above-mentioned generating unit,

[0078] It includes a joining server unit that joins the first grayscale feature vector and the second grayscale feature vector to obtain a multi-partition grayscale distribution feature vector.

[0079] Optionally, the above device,

[0080] A second acquisition module for acquiring the above-mentioned biometric fingerprint sample image and the above-mentioned non-biometric fingerprint sample image - the biometric fingerprint sample image includes images of a plurality of biometric fingerprint samples collected with the green light turned off and the red light and blue light turned on, and the above-mentioned non-biometric fingerprint sample image includes images of a plurality of non-biometric fingerprint samples collected with the green light turned off and the red light and blue light turned on -;

[0081] It further includes a training module obtained by training the fingerprint biometric detection model through supervised learning based on the above-mentioned biometric fingerprint sample image and non-biometric fingerprint sample image.

[0082] Optionally, the above device,

[0083] A third acquisition module for acquiring a fourth original image - the fourth original image is an image of the fingerprint collected with the green light turned on -;

[0084] A recognition module that determines a fingerprint recognition result based on the above-mentioned fourth original image;

[0085] It further includes a second decision module that determines a security verification result based on the fingerprint recognition result and the fingerprint biometric detection result.

[0086] In another embodiment, a fingerprint biometric detection device is provided, wherein the fingerprint biometric detection device comprises an image collector and a processor, and the image collector comprises a fill light and an image sensor;

[0087] The above image collector collects a first original image;

[0088] The above fill illumination fills the image sensor's light by turning off the green light and turning on the red and blue lights during the process of collecting the above first original image;

[0089] The processor processes a first original image to obtain a first fingerprint image and a second fingerprint image, wherein the first fingerprint image is a red channel image and the second fingerprint image is a blue channel image; performs region segmentation for each of the first fingerprint image and the second fingerprint image, wherein the segmented first fingerprint image includes a plurality of first predicted bright areas and a plurality of first predicted dark areas, and the segmented second fingerprint image includes a second predicted bright area and a plurality of second predicted dark areas; determines a multi-partition grayscale distribution feature vector based on the plurality of first predicted bright areas, the plurality of first predicted dark areas, the plurality of second predicted bright areas, and the plurality of second predicted dark areas; and inputs the multi-partition grayscale distribution feature vector into a fingerprint biometric detection model to obtain a fingerprint biometric detection result, wherein the fingerprint biometric detection model is obtained by training through supervised learning based on a biometric fingerprint sample image and a non-biometric fingerprint sample image.

[0090] Optionally, the image collector is further used to collect a fourth original image;

[0091] The above fill illumination is further used to fill the image sensor with light by turning on a green light during the process of collecting the fourth original image;

[0092] The processor is further used to determine a fingerprint recognition result based on the fourth original image and to determine a security verification result based on the fingerprint recognition result and the fingerprint biometric detection result.

[0093] In another embodiment, a computer-readable storage medium is provided, said storage medium has a computer program stored therein, and when said computer program is executed by a processor, the steps of the aforementioned fingerprint biometric detection method are implemented.

[0094] In another embodiment, a computer program product including a command is provided, which, when executed on a computer, causes the computer to perform the steps of the fingerprint biometric detection method described above.

[0095] The technical solution provided in the embodiments of the present application can bring about at least the following beneficial effects:

[0096] In an embodiment of the present application, region segmentation is performed on a first fingerprint image and a second fingerprint image, and a multi-partition grayscale distribution feature vector is determined based on the segmented regions to perform fingerprint biometric detection through a fingerprint biometric detection model, wherein the first fingerprint image is a red channel image and the second fingerprint image is a blue channel image. Since the grayscale value distribution rule of each region of the two-channel images of a non-biometric fingerprint is different from that of a biometric fingerprint, it is possible to recognize whether the grayscale distribution feature of a biometric fingerprint exists in the fingerprint image through the multi-partition grayscale distribution feature vector, and furthermore, to determine whether the said biometric fingerprint is a biometric fingerprint, thereby providing higher safety. Brief explanation of the drawing

[0097] To more clearly explain the technical solution provided in the embodiments of this application, a brief description of the accompanying drawings that need to be used in the embodiments is given below. Clearly, the drawings in the following description are merely some of the one or more embodiments described in this application. To those skilled in the art, other drawings can be obtained based on these drawings without creative effort. FIG. 1 is a schematic diagram of the structure of a fingerprint biometric detection device provided in an embodiment of the present application. FIG. 2 is a flowchart of a fingerprint biometric detection method provided in one embodiment of the present application. FIG. 3 is a schematic diagram of an image sensor used in an embodiment of the present application. FIG. 4 is a schematic diagram of the image data of the extracted red channel provided in an embodiment of the present application. FIG. 5 is a schematic diagram of a first fingerprint image provided in an embodiment of the present application. FIG. 6 is a schematic diagram of the image data of the extracted blue channel provided in an embodiment of the present application. FIG. 7 is a schematic diagram of a second fingerprint image provided in an embodiment of the present application. FIG. 8 is a schematic diagram determining each region of an image provided in an embodiment of the present application. FIG. 9 is a flowchart of another fingerprint biometric detection method provided in an embodiment of the present application. FIG. 10 is a schematic diagram of the structure of a fingerprint biometric detection device provided in an embodiment of the present application. FIG. 11 is a schematic diagram of the structure of another fingerprint biometric detection device provided in an embodiment of the present application. FIG. 12 is a schematic diagram of the structure of a computer device provided in an embodiment of the present application. Specific details for implementing the invention

[0098] In order to clarify the purpose, technical solution, and advantages of the embodiments of the present invention, the embodiments of the present application are described in more detail below in conjunction with the attached drawings.

[0099] Currently, fingerprint recognition is being applied significantly in many scenarios, such as attendance, access control, computer unlocking, mobile phone unlocking, and payments. However, current fingerprint forgery technologies are also continuously evolving; for instance, fingerprint films made of silicone are used to forge another person's fingerprint, successfully bypassing fingerprint recognition to commit illegal acts. To ensure the security of fingerprint recognition, a fingerprint biometric detection function must be added to the fingerprint recognition base to effectively prevent the impersonation of another person's fingerprint using a fingerprint film. For fingerprint images collected in each scenario, fingerprint biometric detection can be performed based on the method provided in the embodiment of this application to obtain fingerprint biometric detection results. Fingerprint recognition is then performed on the collected fingerprint images based on this base to obtain fingerprint recognition results, and if both fingerprint recognition and fingerprint biometric detection results pass, it is determined that the security verification has been passed.

[0100] FIG. 1 is a schematic diagram of the structure of a fingerprint biometric detection device (100) provided in an embodiment of the present application. Referring to FIG. 1, the fingerprint biometric detection device (100) includes an image collector (101) and a processor (102), the image collector (101) includes an LED fill light, a lens, and an image sensor, and the LED fill light includes a green light, a red light, and a blue light.

[0101] The image collector (101) is used to receive a light signal or detect a capacitance, convert the received light signal or the detected capacitance into an electrical signal, and send it to the processor (102). The processor (102) is used to determine the current light intensity change amount or capacitance value based on the received electrical signal. When the light intensity change amount indicates that the current light is significantly dimming or the capacitance value is the capacitance value of a biological finger, a control command to collect a fingerprint image is transmitted to the image collector (101), that is, the image collector (101) is awakened to perform fingerprint collection.

[0102] When the image collector (101) receives a control command, it controls the LED fill light to turn on the green light or turns on the red light and blue light to fill the light based on the fingerprint image collection method provided in the embodiment of the present application, collects the original image through the lens and image sensor, transmits the collected original image to the processor (102), and the processor (102) is equipped with a fingerprint biometric detection model and can obtain a fingerprint biometric detection result by processing the original image based on the fingerprint biometric detection method provided in the embodiment of the present application. Here, the fingerprint biometric detection model is obtained by training through supervised learning based on a biometric fingerprint sample image and a non-biometric fingerprint sample image.

[0103] In an embodiment of the present application, an image collector (101) can collect a first original image, and an LED fill light turns off the green light and turns on the red light and blue light to fill the image sensor with light during the process of collecting the first original image, and a processor (102) processes the first original image to obtain a first fingerprint image and a second fingerprint image, wherein the first fingerprint image is a red channel image and the second fingerprint image is a blue channel. Next, the processor (102) can perform region segmentation for each of the first fingerprint image and the second fingerprint image, wherein the segmented first fingerprint image includes a plurality of first predicted bright areas and a plurality of first predicted dark areas, and the segmented second fingerprint image includes a second predicted bright area and a plurality of second predicted dark areas. The processor (102) can also determine a multi-partition grayscale distribution feature vector based on the plurality of first predicted bright areas, a plurality of first predicted dark areas, a plurality of second predicted bright areas, and a plurality of second predicted dark areas, and then input the multi-partition grayscale feature vector into a fingerprint biometric detection model to obtain a fingerprint biometric detection result.

[0104] Optionally, the image collector (101) may also collect a fourth original image, and the LED fill light may turn on a green light to fill the image sensor with light during the process of collecting the fourth original image, and the processor (102) may determine a fingerprint recognition result based on the fourth original image and determine a security verification result based on the fingerprint recognition result and the fingerprint biometric detection result.

[0105] In an embodiment of the present application, the fingerprint biometric detection device (100) may further include a memory (103), and the memory (103) is used to store data and computer programs necessary for an embodiment of the present application, such as an image containing valid fingerprints used for multiple fingerprint recognitions. A processor (102) may execute a computer program in the memory (103), perform fingerprint recognition by comparing a collected original image with an image of multiple valid fingerprints stored therein to obtain a fingerprint recognition result, and finally, determine a security verification result based on the biometric detection result and the fingerprint recognition result.

[0106] For the specific implementation method to be explained, in which an LED fill light fills the image sensor to collect each original image and a processor (102) processes each original image to obtain a fingerprint biometric detection result or a fingerprint recognition result, refer to the relevant introduction of the method embodiment of FIG. 2 below.

[0107] Optionally, in an embodiment of the present application, the LED fill illumination includes red illumination, green illumination, and blue illumination, and the wavelength ranges of the red illumination, green illumination, and blue illumination corresponding to the three illuminations may be determined based on the Quantum Efficiency (QE) curve of the image sensor. For example, when only the red illumination is turned on, the wavelength corresponding to the peak position when the peak of the QE curve of the image sensor reaches its maximum value may be considered as the center wavelength of the red light, and the range taken one segment to the left and right of the center wavelength may be considered as the wavelength range of the red light. In an embodiment of the present application, the wavelength range of the green light determined based on the method may be 500 nm to 550 nm, the wavelength range of the red light may be 600 nm to 650 nm, and the wavelength range of the blue light may be 420 nm to 480 nm, and these ranges are used only as examples, and the wavelength range finally determined by another device in actual application may be different.

[0108] It should be explained that the fingerprint biometric detection device (100) illustrated in FIG. 1 may be applied to a front-end image collection device, such as a camera, that is, the fingerprint biometric detection method provided in the embodiment of the present application may be applied to a front-end image collection device. In one possible implementation, the function of the fingerprint biometric detection device (100) may be implemented by two independent devices, a front-end device and a back-end device, wherein the back-end device may be connected to the front-end device via wired or wireless connection to perform data transmission. In this case, the front-end device may include the image collector (101) of the aforementioned embodiment to collect original images. Next, the front-end device may transmit the collected original images to the back-end device, and the back-end device may detect and recognize the collected original images through the fingerprint biometric detection method provided in the embodiment of the present application and finally determine the security verification result. Next, the security verification result may be transmitted to the front-end device or another device to notify the user of the security verification pass or fail via text or voice.

[0109] In summary, in an embodiment of the present application, region segmentation can be performed on a first fingerprint image and a second fingerprint image, and a multi-partition grayscale distribution feature vector can be determined based on the segmented regions to perform fingerprint biometric detection through a fingerprint biometric detection model, wherein the first fingerprint image is a red channel image and the second fingerprint image is a blue channel image. Since the distribution rules of grayscale values ​​in the bright and dark regions of the two-channel images of a non-biometric fingerprint differ from the grayscale value distribution of a biometric fingerprint, high safety is achieved by determining whether the fingerprint is a biometric fingerprint by recognizing whether the grayscale distribution features of a biometric fingerprint exist in the fingerprint image through the multi-partition distribution feature vector.

[0110] It should be explained that when the fingerprint detection device provided in the above embodiment detects a fingerprint, although the compartments of the functional modules or components have been described as examples, the functions can be assigned to other functional modules or components and completed based on demand in actual applications; that is, the internal structure of the device can be divided into other functional modules or components to complete all or part of the functions described above. Furthermore, the fingerprint biometric detection device provided in the above embodiment and the fingerprint biometric detection method below belong to the same concept, and since the specific implementation process is described in detail in the method examples below, it will not be repeated here.

[0111] Next, the fingerprint biometric detection method provided in the embodiment of the application will be interpreted and explained in detail.

[0112] FIG. 2 is a flowchart of a fingerprint biometric detection method provided in an embodiment of the present application, and describes the method by applying it to the fingerprint biometric detection device illustrated in FIG. 1 as an example. Referring to FIG. 2, the method comprises the following steps.

[0113] Step 201: Acquire a first fingerprint image and a second fingerprint image. Here, the first fingerprint image is a red channel image and the second fingerprint image is a blue channel image.

[0114] Both the red channel and blue channel images of a biometric fingerprint have alternating bright and dark areas and a relatively regular distribution of grayscale values ​​at pixel points, whereas the two-channel images of a non-biometric fingerprint have a relatively chaotic distribution of grayscale values ​​at pixel points; that is, since the distribution rule of grayscale values ​​corresponding to a non-biometric fingerprint differs from the distribution rule of a biometric fingerprint, fingerprint biometric detection can be performed based on a multi-grayscale distribution rule. In other words, the first fingerprint image obtained in the present embodiment of the application is a red channel image and the second fingerprint image is a blue channel image, and the first fingerprint image and the second fingerprint image refer to images of the same fingerprint.

[0115] As can be seen from the relevant introduction of Fig. 1 mentioned above, the image collector must be woken up before collecting fingerprint images, and it can be determined whether a collection target, such as a finger, is currently detected based on changes in luminosity or capacitance values, and if a collection target is detected, the image collector is woken up and fingerprint images are collected.

[0116] Since the embodiments of the present application must perform fingerprint biometric detection based on red channel images and blue channel images, based on this, the embodiments of the present application provide two image collection methods, namely, two first fingerprint images and two second fingerprint images, and these two implementation methods are introduced below.

[0117] In the first implementation method, the fingerprint biometric detection device can obtain a first original image, and the first original image is a fingerprint image collected when the green light is off and the red light and blue light are on. Next, the fingerprint biometric detection device can generate a first fingerprint image by extracting image data of the red channel of the first original image and generate a second fingerprint image by extracting image data of the blue channel of the first original image.

[0118] In an embodiment of the application, the image sensor includes a red channel, a blue channel, and a green channel. For example, the image sensor may be a Bier mode color image sensor as shown in FIG. 3, where R is the red (R) channel, Gr and Gb are both green (G) channels, and B is the blue (B) channel.

[0119] After activating the image collector, the red and blue lights are turned on simultaneously, and a first original image can be collected through the lens and image sensor; that is, the first original image is the image of the fingerprint collected when the green light is off and the red and blue lights are on. Subsequently, the image collector can transmit the collected original image to the processor. The processor can generate a first fingerprint image by extracting image data from the red channel of the first original image, and generate a second fingerprint image by extracting image data from the blue channel.

[0120] For example, assuming the object to be collected is a biometric fingerprint using the image sensor illustrated in FIG. 3 as an example, the size of the first original image is w0*h0, where w0 and h0 represent the width and height of the first original, respectively, and may also represent the number of pixel points in the width direction and height direction, respectively. The processor can extract image data of the R channel to obtain red channel image data as illustrated in FIG. 4, and generate a first fingerprint image as illustrated in FIG. 5 based on the said red channel image data. The processor can extract image data of the B channel to obtain blue channel image data as illustrated in FIG. 6, and generate a second fingerprint image as illustrated in FIG. 7 based on the said blue channel image data, and the image size of the first fingerprint image and the first fingerprint image are both w0*h0 / 4.

[0121] In the second implementation method, the fingerprint biometric detection device can acquire a second original image and a third original image, the second original image is the fingerprint image collected when the green light and blue light are off and the red light is off, and the third original image is the fingerprint image collected when the green light and red light are off and the blue light is on. Subsequently, the fingerprint biometric detection device can generate a first fingerprint image by extracting image data of the red channel of the second original image and generate a second fingerprint image by extracting image data of the blue channel of the third original image.

[0122] As can be seen, the second implementation method collects the first original image by turning on only the red light, and then collects the second original image by turning on only the blue light; subsequently, it generates the first original image by extracting image data from the red channel of the first original image and generates the second fingerprint image by extracting image data from the blue channel of the second original image. Here, regarding the method of extracting image data, you can refer to the relevant introduction of the first implementation method, which is not repeated here.

[0123] It needs to be explained that in the second implementation method, two original images are collected by turning on only the red light and only the blue light, so the influence of the red light on the blue channel and the influence of the blue light on the red channel can be avoided when the red light and the blue light are turned on simultaneously. However, the first implementation method can reduce the image acquisition time compared to the second implementation method by turning off the green light and turning on the red light and the blue light simultaneously to collect the original images.

[0124] Optionally, since the edges of the collected fingerprint images may be distorted, the fingerprint biometric detection device may extract image data of the red channel and image data of the blue channel, and then perform distortion correction on the image data of the red channel and the image data of the blue channel. Additionally, the fingerprint biometric detection device may crop the red channel image and the blue channel image, respectively, based on preset image size parameters to remove images with unclear edges and obtain a first fingerprint image and a second fingerprint of the same image size.

[0125] Step 202: Region segmentation is performed for each of the first fingerprint image and the second fingerprint image, wherein the segmented first fingerprint image includes a plurality of first predicted bright areas and a plurality of first predicted dark areas, and the segmented second fingerprint image includes a second predicted bright area and a plurality of second predicted dark areas.

[0126] In one embodiment of the present application, a fingerprint biometric detection device divides a first fingerprint image and a second fingerprint image, respectively, based on the distribution rules of bright and dark areas within a red channel image and a blue channel image of a biometric fingerprint, thereby dividing the first fingerprint image into a plurality of first predicted bright areas and a plurality of first predicted dark areas, and dividing the second fingerprint image into a plurality of second predicted bright areas and a plurality of second predicted dark areas.

[0127] From the first fingerprint image shown in FIG. 5 and the second fingerprint image shown in FIG. 7, the distribution rule of bright and dark areas from top to bottom along the image height direction of the red channel image is: bright, dark, bright, dark, bright, and the distribution rule of bright and dark areas from top to bottom along the image height direction of the blue channel image is: dark, bright, dark, bright, dark. That is, both the red channel image and the blue channel image exhibit a distribution rule in which brightness and darkness alternate, and the fingerprint biometric detection device can perform area segmentation on the first fingerprint image and the second fingerprint image based on said distribution rule.

[0128] In an embodiment of the present application, a fingerprint biometric detection device may determine partition parameters based on the distribution rules of bright and dark regions of a red channel image and a blue channel image of a biometric fingerprint, wherein the partition parameters include a plurality of partition width ratios, and then perform region segmentation for each of the first fingerprint image and the second fingerprint image based on the plurality of partition width ratios.

[0129] It needs to be explained that the partition parameter in question may be a parameter pre-set based on the distribution rules of each area. For example, the brightness distribution of the red channel images and blue channel images of multiple biometric fingerprints is statistically analyzed, and based on the brightness distribution situation, the total number of bright and dark areas within the red channel and blue channel images of the fingerprints and the ratio of the image height of the bright and dark areas at the same position from top to bottom are determined, thereby obtaining the partition width ratio for each partition and setting the partition parameter in the fingerprint biometric detection device.

[0130] For example, FIG. 8 is a schematic diagram for determining each region of an image provided in an embodiment of the present application. Referring to FIG. 8, the top-left image of FIG. 8 is a red channel image of a biometric fingerprint, i.e., a first fingerprint image, and the top-right image is a blue channel image of a biometric fingerprint, i.e., a second fingerprint image. Since both the red channel image and the blue channel image exhibit a distribution rule in which brightness and darkness alternate, the red channel image and the blue channel image can be divided into five equal parts from top to bottom. The five regions included in the red channel image are a bright area, a dark area, a bright area, a dark area, and a bright area, and the five regions included in the blue channel image are a dark area, a bright area, a dark area, a bright area, and a dark area from top to bottom. The five regions of the red channel image correspond one-to-one with the five regions of the blue channel, that is, the width ratio of the corresponding regions within the two images is the same. The difference is that the region corresponding to the same height range within the blue channel image as the bright area of ​​the red channel image is the dark area. When the distribution of brightness areas of the red channel image and the blue channel image is combined, it can be seen that both the blue channel image and the red channel image can be divided into five regions as shown in FIG. 8 based on the same division rule, respectively, region 1, region 2, region 3, region 4, and region 5, and the partition width ratios corresponding to the five regions are shown in FIG. 8, for example, 20 / 66, 10 / 66, 15 / 66, 9 / 66, and 12 / 66 from top to bottom, respectively, and other ratios are also possible.

[0131] A fingerprint biometric detection device can divide a first fingerprint image and a second fingerprint image, respectively, according to the partition width ratio shown in FIG. 8, and the divided first fingerprint image includes three first predicted bright areas and two first predicted dark areas, wherein the three first predicted bright areas correspond to areas 1, 3, and 5 in FIG. 8, and the two first predicted dark areas correspond to areas 2 and 4 in FIG. 8. Similarly, the divided second fingerprint image includes two second predicted bright areas and three second predicted dark areas, wherein the two second predicted bright areas correspond to areas 2 and 4 in FIG. 8, and the three second predicted dark areas correspond to areas 1, 3, and 5 in FIG. 8.

[0132] Optionally, if the image sizes of the first fingerprint image and the second fingerprint image are preset image sizes, the partition parameter may include multiple width values. For example, assuming the height of the preset image size is 66 pixels, the width values ​​of the five regions shown in FIG. 8 are 20, 10, 15, 9, and 12 pixels in order from top to bottom. The fingerprint biometric detection device can perform region division for the first fingerprint image and the second fingerprint image based on the multiple width values.

[0133] Step 203: Determine a multi-partition grayscale distribution feature vector based on a plurality of first predicted bright areas, a plurality of first predicted dark areas, a plurality of second predicted bright areas, and a plurality of second predicted dark areas.

[0134] In an embodiment of the present application, the image sizes of the first fingerprint image and the second fingerprint image are the same, and after performing region segmentation for each of the first fingerprint image and the second fingerprint image, the fingerprint biometric detection device determines a multi-partition grayscale distribution feature vector based on the grayscale values ​​of each statistically segmented region.

[0135] In an embodiment of the present application, a fingerprint biometric detection device can determine a first grayscale feature vector based on a plurality of first predicted bright regions and a plurality of second predicted dark regions, and can determine a second grayscale feature vector based on said plurality of first predicted dark regions and said plurality of second predicted bright regions. Next, the fingerprint biometric detection device can generate a multi-partition grayscale distribution feature vector based on the first grayscale feature vector and the second grayscale feature vector.

[0136] A fingerprint biometric detection device can determine a first grayscale feature vector and a second grayscale feature vector by statistically analyzing the grayscale values ​​of each region of a first fingerprint image and a second fingerprint image.

[0137] In one embodiment of the present application, a fingerprint biometric detection device can determine the number of pixel points corresponding to each grayscale value within a plurality of first predicted bright areas and a plurality of second predicted dark areas, and generate a first grayscale feature vector based on the number of pixel points corresponding to each grayscale value within the plurality of first predicted bright areas and a plurality of second predicted dark areas. That is, the fingerprint biometric detection device can determine the first grayscale feature vector by statistically analyzing the number of pixel points corresponding to each grayscale value within a plurality of first predicted bright areas of a first fingerprint image and a plurality of second predicted dark areas of a second fingerprint image. Here, the plurality of first predicted bright areas and the plurality of second predicted dark areas correspond one-to-one, and the corresponding areas are areas within the same height range of the two images.

[0138] For example, taking the schematic diagram of the partition introduced in FIG. 8 as an example, the fingerprint biometric detection device can statistically determine the number of pixel points corresponding to each grayscale value from '0' to '255' within regions 1, 3, and 5 of the first fingerprint image, and obtain vector histr1 by arranging the number of pixel points corresponding to each grayscale value obtained through statistical determination in the order of grayscale values ​​from 0 to 255. Then, the number of pixel points for each grayscale value from '0' to '255' within regions 1, 3, and 5 of the second fingerprint image is statistically determined, and vector histb1 is obtained by arranging the grayscale values ​​in the order of 0 to 255. Here, the lengths of both vector histr1 and vector histb1 are 256. Next, vector hist1 and vector histb1 can be added to obtain the first grayscale feature vector hisfp1, and the length of vector hisfp1 is also 256.

[0139] Similarly, the fingerprint biometric detection device can determine the number of pixel points corresponding to each grayscale value within the plurality of first predicted dark regions and the plurality of second predicted bright regions, and generates a second grayscale feature vector based on the number of pixel points corresponding to each grayscale value within the plurality of first predicted dark regions and the plurality of second predicted bright regions. That is, the fingerprint biometric detection device determines the second grayscale feature vector by statistically analyzing the number of pixel points corresponding to each grayscale value within the plurality of first predicted dark regions of the first fingerprint image and the second predicted bright regions of the second fingerprint image.

[0140] For example, taking the schematic diagram of the partition introduced in FIG. 8 as an example, the fingerprint biometric detection device can statistically analyze pixel points for each grayscale value from '0' to '255' within regions 2 and 4, and obtain vector histr2 by arranging the number of pixel points corresponding to each grayscale value obtained through statistical analysis in the order of grayscale values ​​0 to 255. Then, the number of pixel points for each grayscale value from '0' to '255' within regions 2 and 4 of the second fingerprint image is statistically analyzed, and the number of pixel points corresponding to each grayscale value obtained through statistical analysis is arranged in the order of grayscale values ​​0 to 255 to obtain vector histb2. Here, the lengths of both vector histr2 and vector histb2 are 256. Next, vector histr2 and vector histb2 can be added to obtain the second grayscale feature vector hisfp2, and the length of vector hisfp2 is also 256.

[0141] Optionally, since the exposure when the fingerprint biometric detection device collects an image affects the collected image—whereby if the exposure is too high, the collected image becomes too bright, and if the exposure is too low, the collected image becomes too dark—the grayscale range can be narrowed to reduce the influence of exposure and retain valid grayscale information; that is, a single grayscale value statistical range can be determined based on the exposure parameters of the image collector collecting the image of the fingerprint. In actual applications, the exposure parameters of different image collection devices may differ, and a corresponding grayscale value statistical range can be set based on the exposure parameters of each device. It should be noted that in the embodiments of the present application, the fingerprint biometric detection device is an image collection device. Furthermore, by narrowing the grayscale value statistical range, the statistical amount can be reduced and the detection speed can be increased.

[0142] Based on the above introduction, in this embodiment of the present application, a fingerprint biometric detection device can acquire a grayscale value statistical range, determine the number of pixel points corresponding to each grayscale value within the grayscale value statistical range within a plurality of first predicted bright areas and a plurality of second predicted dark areas based on the grayscale value statistical range, and generate a first grayscale feature vector based on the number of pixel points corresponding to each grayscale value within the grayscale value statistical range within the plurality of first predicted bright areas and a plurality of second predicted dark areas.

[0143] Likewise, the fingerprint biometric detection device can determine the number of pixel points corresponding to each grayscale value within the grayscale value statistical range within the plurality of first predicted dark areas and the plurality of second predicted bright areas based on the grayscale value statistical range, and generates a second grayscale feature vector based on the number of pixel points corresponding to each grayscale value within the grayscale value statistical range within the plurality of first predicted dark areas and the plurality of second predicted bright areas.

[0144] That is, the fingerprint biometric detection device can statistically determine only the number of pixel points corresponding to each grayscale value within the corresponding grayscale value statistical range in each area of ​​the first fingerprint image and the second fingerprint image, and can generate the first grayscale feature vector and the second grayscale feature vector according to the method described above.

[0145] For example, since the range of grayscale value statistics can be from '11' to '230', the lengths of the first grayscale feature vector and the second grayscale feature vector obtained based on the above-described method are both 220, and the length of the vectors is reduced so that the amount of subsequent statistics can be reduced.

[0146] After acquiring the first grayscale feature vector and the second grayscale feature vector, the fingerprint biometric detection device can acquire a multi-partition grayscale distribution feature vector by joining the first grayscale feature vector and the second grayscale feature vector.

[0147] For example, the first grayscale feature vector is hisfp1 and the second grayscale feature vector is hisfp2, and after joining the first grayscale feature vector and the second grayscale feature vector, the obtained multi-partition grayscale distribution feature vector may be [hisfp1, hisfp2].

[0148] The step of obtaining a multi-partition grayscale distribution feature vector by the fingerprint biometric detection device described above is as follows: First, two vectors are obtained by statistically analyzing a plurality of first predicted bright areas of a first fingerprint image and a plurality of second predicted dark areas of a second fingerprint image, respectively, and these two vectors are added to obtain a first grayscale feature vector. Then, two vectors are obtained by statistically analyzing a plurality of first predicted dark areas of a first fingerprint image and a plurality of second predicted bright areas of a second fingerprint image, respectively, and these two vectors are added to obtain a second grayscale feature vector. Then, the first grayscale feature vector and the second grayscale feature vector are joined to obtain a multi-partition grayscale distribution feature vector.

[0149] Optionally, in an embodiment of the present application, the fingerprint biometric detection device also first obtains a vector histr1 by statistically analyzing a plurality of first predicted bright regions of a first fingerprint image, obtains a vector histr2 by statistically analyzing a plurality of first predicted dark regions of the first fingerprint image, obtains a first grayscale junction vector histr by concatenating vector hist1 and vector histr2, then obtains a vector histb1 by statistically analyzing a plurality of second predicted dark regions of a second fingerprint image, obtains a vector histb2 by statistically analyzing a plurality of second predicted bright regions of the second fingerprint image, obtains a second grayscale junction vector histb by concatenating vector histb1 and vector histb2, and then obtains a multi-partition grayscale distribution feature vector hisfp by adding vector histr and vector histb. That is, after obtaining grayscale junction vectors for bright and dark regions corresponding to each of the two images by statistically analyzing two images separately, the multi-partition grayscale distribution feature vector is obtained by adding the two grayscale junction vectors. In this way, the fingerprint biometric detection device can process one image first and then another image without alternating between changing the image data of the images to be statistically analyzed.

[0150] It needs to be explained that in the step of determining the multi-partition grayscale distribution feature vector by statistically analyzing the grayscale values ​​of each region mentioned above, the statistical order of each region may not be in a specific order, and the statistical order can be determined based on the actual situation.

[0151] Optionally, the fingerprint biometric detection device can also display the above statistical results as a grayscale histogram and determine each vector based on the grayscale histogram.

[0152] Step 204: Input a multi-segmented grayscale distribution feature vector into a fingerprint biometric detection model to obtain a fingerprint biometric detection result.

[0153] In an embodiment of the present application, a fingerprint biometric detection device can obtain a fingerprint biometric detection result by acquiring a multi-partition grayscale distribution feature vector and then inputting the vector into a fingerprint biometric detection model.

[0154] Optionally, the fingerprint biometric detection model may be a Support Vector Machine (SVM) classifier, a neural network model, a Random Forest model, etc., and the embodiments of the present application are not limited thereto. The fingerprint biometric detection result may be 'biometric fingerprint' or 'non-biometric fingerprint' or '0' or '1', where '0' means not a biometric fingerprint and '1' means a biometric fingerprint.

[0155] It should be explained that, in the embodiments of the present application, a fingerprint biometric detection model may be obtained by training through supervised learning based on biometric fingerprint sample images and non-biometric fingerprint sample images, and the fingerprint biometric detection model is a pre-trained model, and may be obtained by training the fingerprint biometric detection model through a fingerprint biometric detection device or through other devices, such as a background device or other computer device, and then the fingerprint biometric detection model is placed in the fingerprint biometric detection device. The embodiments of the present application introduce a fingerprint biometric detection model obtained through training by a fingerprint biometric detection device as an example.

[0156] In an embodiment of the present application, a fingerprint biometric detection device can acquire a biometric fingerprint sample image and a non-biometric fingerprint sample image, wherein the biometric fingerprint sample image includes images of a plurality of biometric fingerprint samples collected with the green light turned off and the red light and blue light turned on. The non-biometric fingerprint sample image includes images of a plurality of non-biometric fingerprint samples collected with the green light turned off and the red light and blue light turned on. Subsequently, the fingerprint biometric detection device can acquire a fingerprint biometric detection model by training it through supervised learning based on the biometric fingerprint sample image and the non-biometric fingerprint sample image.

[0157] It should be explained that the method for acquiring the biometric fingerprint sample image and the non-biometric fingerprint sample image is the same as the method for acquiring the first original image described above, and a detailed explanation is omitted here.

[0158] After acquiring biometric fingerprint sample images and non-biometric fingerprint sample images, the fingerprint biometric detection device can acquire a multi-partition grayscale distribution feature vector corresponding to each sample image according to the method for acquiring multi-partition grayscale distribution features described above. Subsequently, a fingerprint biometric detection model can be trained and obtained based on the multi-partition grayscale distribution feature vector corresponding to each sample image.

[0159] Optionally, each sample image corresponds to one sample label, the sample label of the biometric fingerprint sample image may be '1' and the sample label of the non-biometric fingerprint sample image may be '0', and the fingerprint biometric detection device is obtained by training a fingerprint biometric detection model through supervised learning based on a multi-partition grayscale distribution feature vector and sample label corresponding to each sample image.

[0160] For example, assuming the fingerprint biometric detection model is an SVM classifier, the fingerprint biometric detection device can be obtained by training the SVM classifier through supervised learning based on a multi-partition grayscale distribution feature vector and sample labels corresponding to each sample image, and said SVM classifier is a single binary classifier.

[0161] FIG. 9 is a flowchart of another fingerprint biometric detection method provided in an embodiment of the present application. Referring to FIG. 9, assuming the fingerprint biometric detection model is an SVM classifier, feature extraction is performed on each of the first fingerprint image (ImgR) and the second fingerprint image (ImgB) to obtain vectors histr and histb, and then feature synthesis is performed, that is, by joining vectors histr and histb to obtain a multi-partition grayscale distribution feature vector hiafp, and the vector hisfp is input into the SVM classifier to output a fingerprint biometric detection result.

[0162] The fingerprint biometric detection method provided in the embodiments of the present application has been described above. In actual applications, it is necessary to determine the final security verification result by combining fingerprint recognition and fingerprint biometric detection. That is, if the fingerprint recognition result indicates that the collected fingerprint is a valid fingerprint and the fingerprint biometric detection result indicates that the collected fingerprint is a biometric fingerprint, it is determined that the security verification has been passed. Based on this, the embodiments of the present application also provide a fingerprint security verification method to be introduced below.

[0163] In an embodiment of the present application, the fingerprint biometric detection device may also acquire a fourth original image, which is an image of the fingerprint collected while the green light is turned on, and determine a fingerprint recognition result based on the fourth original image. Subsequently, the fingerprint biometric detection device may determine a security verification result based on the fingerprint recognition result and the fingerprint biometric detection result.

[0164] The green channel image of a biometric fingerprint is more suitable for fingerprint recognition than the red channel image and the blue channel image because it lacks a relatively clear fingerprint pattern, that is, a feature of alternating light and dark. Additionally, the biometric fingerprint image collected with all three colors of light turned on also exhibits a very clear fingerprint pattern, making it suitable for fingerprint recognition. Based on this, if the image collector is activated to perform fingerprint collection, not only is the original image of the fingerprint in question collected when the red and blue lights are turned on, but the image of the fingerprint in question is also collected when the green light is turned on to obtain a fourth original image. Here, the state where the green light is turned on may mean a state where the green light is turned on and the red and blue lights are turned off, or a state where the green, red, and blue lights are all turned on. In other words, after activating the image collector, one collection can be completed by collecting the first original image and the fourth original image, or one collection can be completed by collecting the second original image, the third original image, and the fourth original image.

[0165] Additionally, all lights may be turned off after each collection is completed, and the embodiments of the present application do not limit the order of the original images to be collected. For example, the fourth original image may be collected by first turning on only the green light, then turning off the green light and turning on the red and blue lights to collect the first original image. Also, for example, the fourth original image may be collected by first turning on the green light, red light, and blue light, and then the first original image may be collected while turning off the green light and keeping the red and blue lights on. Again, for example, the second original image may be collected by first turning on the red light, then turning off the red light and turning on the blue light to collect the third original image, and finally turning off the blue light and turning on the green light to collect the fourth original image.

[0166] After acquiring the fourth original image, the fingerprint biometric detection device extracts image data of the green channel of the fourth original image to generate a third fingerprint image, then extracts fingerprint features of the third fingerprint image, such as the number of lines, line shape, key points, breakpoints, intersections, direction, curvature, etc., and obtains a fingerprint recognition result by comparing the extracted fingerprint features with features corresponding to each stored valid fingerprint image. Here, the fingerprint recognition result may be a 'valid fingerprint' or an 'invalid fingerprint'.

[0167] Here, the implementation method for extracting image data of the green channel of the fourth original image and generating the third fingerprint image may refer to the introduction to the generation of the first fingerprint image and the second fingerprint image described above.

[0168] For example, taking the color image sensor in the Bier mode illustrated in FIG. 3 as an example, the fingerprint biometric detection device may obtain a third fingerprint image with an image size of w0*h0 / 4 by acquiring a fourth original image and then extracting image data of the Gr channel or image data of the Gb channel. Optionally, the fingerprint biometric detection device may perform distortion correction on the extracted image to obtain a corrected third fingerprint image. Optionally, furthermore, the fingerprint biometric detection device may also crop the corrected image based on preset image size parameters to obtain a clearer middle portion, which is used as the third fingerprint image.

[0169] It should be noted that the image size of the third fingerprint image may be the same as or different from the image size of the first fingerprint image. If they are the same, you simply need to set a set of preset image size parameters.

[0170] Optionally, the fingerprint biometric detection device may obtain a fingerprint recognition result by directly extracting fingerprint features of the fourth original image after acquiring the fourth original image and comparing the extracted fingerprint features with features corresponding to the stored images of each valid fingerprint.

[0171] After obtaining fingerprint recognition results and fingerprint biometric detection results based on the above method, the fingerprint biometric detection device can determine the final security verification result based on these two results. For example, if the fingerprint recognition result indicates that the collected fingerprint is a valid fingerprint and the fingerprint biometric detection result indicates that the collected fingerprint is a biometric fingerprint, it is determined that the security verification has passed. If the fingerprint recognition result indicates that the collected fingerprint is not a valid fingerprint or the fingerprint biometric detection result indicates that the collected fingerprint is not a biometric fingerprint, it is determined that the security verification has failed, and the fingerprint biometric detection device may notify the user via a text prompt or voice.

[0172] In summary, in an embodiment of the present application, region segmentation is performed on a first fingerprint image and a second fingerprint image, and a multi-partition grayscale distribution feature vector is determined based on the segmented regions to perform fingerprint biometric detection through a fingerprint biometric detection model, wherein the first fingerprint image is a red channel image and the second fingerprint image is a blue channel image. Since the grayscale value distribution rule of each region of the two-channel images of a non-biometric fingerprint is different from that of a biometric fingerprint, it is possible to recognize whether the grayscale distribution feature of a biometric fingerprint exists in the fingerprint image through the multi-partition grayscale distribution feature vector, and furthermore, to determine whether the said biometric fingerprint is a biometric fingerprint, thereby providing higher safety.

[0173] FIG. 10 is a schematic diagram of the structure of a fingerprint biometric detection device (1000) provided in one embodiment of the present application, wherein the fingerprint biometric detection device (1000) may be implemented as part or all of a computer device by software, hardware, or a combination thereof. Referring to FIG. 10, the device includes a first acquisition module (1001), a partition module (1002), a first determination module (1003), and a detection module (1004).

[0174] The first acquisition module (1001) is used to acquire a first fingerprint image and a second fingerprint image, and the first fingerprint image and the second fingerprint image are images of the same fingerprint, the first fingerprint image is a red channel image, and the second fingerprint image is a blue channel image;

[0175] The partition module (1002) performs region division for each of the first fingerprint image and the second fingerprint image, wherein the divided first fingerprint image includes a plurality of first predicted bright areas and a plurality of first predicted dark areas, and the divided second fingerprint image includes a second predicted bright area and a plurality of second predicted dark areas;

[0176] The first determination module (1003) is used to determine a multi-partition grayscale distribution feature vector based on a plurality of first predicted bright areas, a plurality of first predicted dark areas, a plurality of second predicted bright areas and a plurality of predicted dark areas;

[0177] The detection module (1004) is used to obtain fingerprint biometric detection results by inputting a multi-partition grayscale distribution feature vector into a fingerprint biometric detection model. The fingerprint biometric detection model is obtained by training through supervised learning based on a biometric fingerprint sample image and a non-biometric fingerprint sample image.

[0178] Optionally, the first acquisition module (1001) is,

[0179] A first acquisition unit for acquiring a first original image - the first original image is an image of a fingerprint collected while the green light is off and the red and blue lights are on -;

[0180] A first extraction unit for generating a first fingerprint image by extracting image data of the red channel of a first original image;

[0181] It includes a second extraction unit for generating a second fingerprint image by extracting image data of the blue channel of a first original image.

[0182] Optionally, the first acquisition module (1001) is,

[0183] A second acquisition unit for acquiring a second original image and a third original image - the second original image is an image of a fingerprint collected with the green and blue lights off and the red light on, and the third original image is a fingerprint image collected with the green and red lights off and the blue light on -;

[0184] A third extraction unit for generating a first fingerprint image by extracting image data of the red channel of a second original image;

[0185] It includes a fourth extraction unit for generating a second fingerprint image by extracting image data of the blue channel of the third original image.

[0186] Optionally, the partition module (1002) is,

[0187] A first determination unit for determining partition parameters based on the distribution rules of bright and dark regions of the red channel image and blue channel image of a biometric fingerprint - the partition parameters include a plurality of partition width ratios -;

[0188] It includes a partitioning unit for performing area partitioning for each of the first fingerprint image and the second fingerprint image based on a plurality of partition width ratios.

[0189] Optionally, the image sizes of the first fingerprint image and the second fingerprint image are the same, and the first determination module (1003) is,

[0190] A second determination unit for determining a first grayscale feature vector based on a plurality of first predicted bright regions and a plurality of second predicted dark regions;

[0191] A third determination unit for determining a second grayscale feature vector based on a plurality of first predicted dark regions and a plurality of second predicted bright regions;

[0192] It includes a generation unit for generating a multi-partition grayscale distribution feature vector based on a first grayscale feature vector and a second grayscale feature vector.

[0193] Optionally, the second decision unit,

[0194] A first determination sub-unit for determining the number of pixel points corresponding to each grayscale value within a plurality of first predicted bright regions and a plurality of second predicted dark regions;

[0195] It includes a first generating sub-unit for generating a first grayscale feature vector based on the number of pixel points corresponding to each grayscale value within a plurality of first predicted bright regions and a plurality of second predicted dark regions.

[0196] Optionally, the third decision unit,

[0197] A second determination sub-unit for determining the number of pixel points corresponding to each grayscale value within a plurality of first predicted dark regions and a plurality of second predicted bright regions;

[0198] It includes a second generating sub-unit for generating a second grayscale feature vector based on the number of pixel points corresponding to each grayscale value within a plurality of first predicted dark regions and a plurality of second predicted bright regions.

[0199] Optionally, the first decision module (1003) is,

[0200] It further includes a third acquisition unit for acquiring a grayscale value statistical range, wherein the grayscale value statistical range is determined based on the exposure parameters of an image acquisition device that collects an image of a fingerprint;

[0201] The second decision unit is,

[0202] A third determination sub-unit for determining the number of pixel points corresponding to each grayscale value within the grayscale value statistical range in a plurality of first predicted bright areas and a plurality of second predicted dark areas, based on the grayscale value statistical range;

[0203] It includes a third generating sub-unit for generating a first grayscale feature vector based on the number of pixel points corresponding to each grayscale value within a grayscale value statistical range in a plurality of first predicted bright regions and a plurality of second predicted dark regions.

[0204] Optionally, the generation unit,

[0205] It includes a joining sub-unit for obtaining a multi-partition grayscale distribution feature vector by joining a first grayscale feature vector and a second grayscale feature vector.

[0206] Optionally, the device (1000) is,

[0207] A second acquisition module for acquiring biometric fingerprint sample images and non-biometric fingerprint sample images - the biometric fingerprint sample images include a plurality of biometric fingerprint sample images collected with the green light off and the red light and blue light on, and the non-biometric fingerprint sample images include images of a plurality of non-biometric fingerprint samples collected with the green light off and the red light and blue light on -;

[0208] It further includes a training model to be obtained by training a fingerprint biometric detection model through supervised learning based on biometric fingerprint sample images and non-biometric fingerprint sample images.

[0209] Optionally, referring to FIG. 11, the device (1000) is,

[0210] A third acquisition module (1005) for acquiring a fourth original image - the fourth original image is an image of a fingerprint collected with the green light turned on -;

[0211] A recognition module (1006) for determining a fingerprint recognition result based on a fourth original image;

[0212] It further includes a second decision module (1007) for determining a security verification result based on fingerprint recognition results and fingerprint biometric detection results.

[0213] In summary, in an embodiment of the present application, region segmentation can be performed on a first fingerprint image and a second fingerprint image, and then a multi-grayscale distribution feature vector is determined based on the segmented regions to perform fingerprint biometric detection through a fingerprint biometric detection model, wherein the first fingerprint image is a red channel image and the second fingerprint image is a blue channel image. Since the distribution rules of grayscale values ​​in the bright and dark regions of the two-channel images of a non-biometric fingerprint differ from those of a biometric fingerprint, safety is enhanced by recognizing whether the grayscale distribution features of a biometric fingerprint exist in the fingerprint image through the multi-partition grayscale distribution feature vector, and further determining whether the fingerprint is a biometric fingerprint.

[0214] It should be explained that when the fingerprint detection device provided in the above embodiment detects a fingerprint, although it was described by way of example through the partitioning of each functional module, in actual applications, the function can be completed by assigning it to other functional modules based on demand; that is, the internal structure of the device can be divided into different functional modules to complete all or part of the aforementioned function. Furthermore, the fingerprint biometric detection device and fingerprint biometric detection method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process refers to the method embodiment and is not repeated here.

[0215] FIG. 12 is a structural block diagram of a computer device (1200) provided by an embodiment of the present application. The computer device (1200) may be a fingerprint biometric detection device capable of collecting fingerprints and performing security verification, such as a smartphone, tablet computer, laptop computer, or desktop computer.

[0216] Generally, the computer device (1200) includes a processor (1201) and memory (1202).

[0217] The processor (1201) may include one or more processing cores, such as a quad-core processor or a dodeca-core processor. The processor (1201) may be realized by adopting at least one hardware form among a Digital Signal Processing (DSP), a Field Programmable Gate Array (FPGA), and a Programmable Logic Array (PLA). The processor (1201) may further include a main processor and an auxiliary processor, the main processor is for processing data in a wake-up state and is also referred to as a Central Processing Unit (CPU); and the auxiliary processor is a low-power processor for processing data in a standby state. In some embodiments, a Graphics Processing Unit (GPU) may be integrated into the processor (1201), and the GPU is used to render and produce content to be displayed on a display screen. In some embodiments, the processor (1201) may further include an Artificial Intelligence (AI) processor, said AI processor is used to process computing operations related to machine learning.

[0218] The memory (1202) may include one or more non-transient computer-readable storage media. The memory (1202) may further include high-speed random access memory and one or more non-volatile memory such as disk storage devices, flash storage devices, etc. In some embodiments, the non-transient computer-readable storage media of the memory (1202) is used to store at least one instruction, said at least one instruction is executed by the processor (1201) and used to realize the fingerprint biometric detection method provided in the method embodiment of the embodiment of the present application.

[0219] In some embodiments, the computer device (1200) may optionally include a peripheral interface (1203) and at least one peripheral. The processor (1201), memory (1202), and peripheral interface (1203) may be connected via a bus or a signal line. Each peripheral may be connected to the peripheral interface (1203) via a bus, a signal line, or a circuit board. Specifically, the peripheral includes at least one of a radio frequency circuit (1204), a display screen (1205), a camera assembly (1206), an audio circuit (1207), a positioning assembly (1208), and a power supply (1209).

[0220] The peripheral interface (1203) may be used to connect at least one peripheral device associated with I / O (Input / Output) to the processor (1201) and memory (1202). In some embodiments, the processor (1201), memory (1202), and peripheral interface (1203) are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor (1201), memory (1202), and peripheral interface (1203) may be implemented on separate chips or circuit boards. The present embodiment is not limited thereto.

[0221] A radio frequency circuit (1204) is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The radio frequency circuit (1204) communicates with a communication network and other communication devices through electromagnetic signals. The radio frequency circuit (1204) converts electrical signals into electromagnetic signals for transmission or converts received electromagnetic signals into electrical signals. Optionally, the radio frequency circuit (1204) includes an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a subscriber identification module card, etc. The radio frequency circuit (1204) may communicate with other computer devices through at least one wireless communication protocol. Such wireless communication protocols include, but are not limited to, the World Wide Web, Metropolitan Area Network, intranet, various generations of mobile communication networks (2G, 3G, 4G, and 5G), short-range wireless communication networks and / or WiFi (Wireless Fidelity) networks. In some embodiments, the radio frequency circuit (1204) may further include a circuit related to Near Field Communication (NFC), and embodiments of the present application are not limited thereto.

[0222] A display screen (1205) is used to display a UI (User Interface). The UI may include graphics, text, icons, videos, and combinations thereof. When the display screen (1205) is a touch display screen, the display screen (1205) also has the ability to collect touch signals on or above the surface of the display screen (1205). Such touch signals may be input to a processor (1201) as control signals and processed. In this case, the display screen (1205) may also be used to provide virtual buttons and / or a virtual keyboard, i.e., also referred to as soft buttons and / or a soft keyboard. In some embodiments, the display screen (1205) may be one and may be installed on the front panel of the computer device (1200); in other embodiments, the display screen (1205) may be at least two and each may be installed on different surfaces of the computer device (1200) or designed to be foldable; In some other embodiments, the display screen (1205) may be a flexible screen and may be placed on a curved or folded surface of the computer device (1200). Furthermore, the display screen (1205) may also be installed as a screen of a special shape, that is, an irregular shape rather than a rectangle. The display screen (1205) may be made of a material such as an LCD (Liquid Crystal Display) or an OLED (Organic Light-Emitting Diode).

[0223] The camera assembly (1206) is used to collect images or videos. Optionally, the camera assembly (1206) includes a front camera and a rear camera. Typically, the front camera is placed on the front panel of the computer device and the rear camera is placed on the rear of the computer device. In some embodiments, the rear camera is at least two, each being one of a main camera, a depth-of-field camera, a wide-angle camera, and a telephoto camera, and the fusion of the main camera and the depth-of-field camera is used to realize a background blur function, and the fusion of the main camera and the wide-angle camera is used to implement panoramic shooting and VR (Virtual Reality) shooting functions or other fused shooting functions. In some embodiments, the camera assembly (1206) may also include a flash. The flash may be a single color temperature flash or a dual color temperature flash. A dual color temperature flash refers to a combination of a warm flash and a cool flash and can be used to compensate for light at different color temperatures.

[0224] The audio circuit (1207) may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, convert the sound waves into electrical signals, and input them to the processor (1201) for processing or to input them to the radio frequency circuit (1204) to enable voice communication. For the purpose of stereo acquisition or noise reduction, there may be multiple microphones, each of which may be placed in different parts of the computer device (1200). The microphone may be an array microphone or an omnidirectional collection microphone. The speaker is used to convert electrical signals coming from the processor (1201) or the radio frequency circuit (1204) into sound waves. The speaker may be a traditional thin-film speaker or a piezoelectric ceramic speaker. If the speaker is a piezoelectric ceramic speaker, it may not only convert electrical signals into sound waves that can be heard by humans, but also convert electrical signals into sound waves that cannot be heard by humans for use in distance measurement applications. In some embodiments, the audio circuit (1207) may also include a headphone jack.

[0225] The positioning component (1208) positions the current geographical location of the computer device (1200) to implement navigation or LBS (Location Based Service). The positioning component (1208) may be a positioning component based on the US Global Positioning System (GPS), China's Beidou system, or Russia's Galileo system.

[0226] The power source (1209) is used to supply power to each component of the computer device (1200). The power source (1209) may be an alternating current, direct current, disposable battery, or rechargeable battery. If the power source (1209) includes a rechargeable battery, the rechargeable battery may be a wired rechargeable battery or a wireless rechargeable battery. A wired rechargeable battery is a battery that is charged via a wired line, and a wireless rechargeable battery is a battery that is charged via a wireless coil. The rechargeable battery may also be used to support high-speed charging technology.

[0227] In some embodiments, the computer device (1200) further includes one or more sensors (1210). The one or more sensors (1210) include, but are not limited to, an accelerometer (1211), a gyroscope (1212), a pressure sensor (1213), a fingerprint sensor (1214), an optical sensor (1215), and a proximity sensor (1216).

[0228] The acceleration sensor (1211) can detect the magnitude of acceleration in three coordinate axes of a coordinate system constructed by the computer device (1200). For example, the acceleration sensor (1211) can be used to detect the components of gravitational acceleration in three coordinate axes. The processor (1201) can control the display screen (1205) to display the user interface in a horizontal view or a vertical view based on the gravitational acceleration signal collected by the acceleration sensor (1211). The acceleration sensor (1211) can also be used for collecting game or user movement data.

[0229] The gyro sensor (1212) can detect the body direction and rotation angle of the computer device (1200), and the gyro sensor (1212) can collect 3D movements of the user on the computer device (1200) in cooperation with the accelerometer (1211). The processor (1201) can implement functions such as motion detection (e.g., changing the UI based on the user's tilt operation), image stabilization during shooting, game control, and inertial navigation based on the data collected by the gyro sensor (1212).

[0230] A pressure sensor (1213) may be placed on the side bezel frame of the computer device (1200) and / or on the lower layer of the display screen (1205). When the pressure sensor (1213) is placed on the side bezel frame of the computer device (1200), it can detect a user's grip signal on the computer device (1200), and the processor (1201) performs left-hand and right-hand recognition or shortcut key operation based on the grip signal collected by the pressure sensor (1213). When the pressure sensor (1213) is placed on the lower layer of the display screen (1205), the processor (1201) performs control on operable controls on the UI interface based on the user's pressure operation on the display screen (1205). Operable controls include at least one of button controls, scroll bar controls, icon controls, and menu controls.

[0231] A fingerprint sensor (1214) is used to collect a user's fingerprint, and a processor (1201) recognizes the user's identity based on the fingerprint collected by the fingerprint sensor (1214), or the fingerprint sensor (1214) recognizes the user's identity based on the collected fingerprint. When the user's identity is recognized as a trusted identity, the processor (1201) authorizes the user to perform related sensitive operations, including unlocking the screen, viewing encrypted information, downloading software, making payments, and changing settings. The fingerprint sensor (1214) may be installed on the front, back, or side of the computer device (1200). If the computer device (1200) is provided with a physical button or a manufacturer's logo, the fingerprint sensor (1214) may be integrated into the physical button or the manufacturer's logo.

[0232] A light sensor (1215) is used to collect the intensity of ambient light. In one embodiment, the processor (1201) can control the display brightness of the display screen (1205) based on the intensity of ambient light collected by the optical sensor (1215). Specifically, if the intensity of ambient light is high, the display brightness of the display screen (1205) is increased; and if the intensity of ambient light is low, the display brightness of the display screen (1205) is decreased. In another embodiment, the processor (1201) can also dynamically adjust the shooting parameters of the camera assembly (1206) based on the intensity of ambient light collected by the optical sensor (1215).

[0233] The proximity sensor (1216), also called a distance sensor, is generally installed on the front panel of the computer device (1200). The proximity sensor (1216) is used to collect the distance between the user and the front of the computer device (1200). In one embodiment, when the proximity sensor (1216) detects that the distance between the user and the front of the computer device (1200) is gradually decreasing, the processor (1201) controls the display screen (1205) to switch from a bright screen state to a resting state; and when the proximity sensor (1216) detects that the distance between the user and the front of the computer device (1200) is gradually increasing, the processor (1201) controls the display screen (1205) to switch from a resting state to a bright screen state.

[0234] A person skilled in the art will understand that the structure illustrated in FIG. 12 does not constitute a limitation for the computer device (1200), and may include more or fewer components than illustrated, or combine some components, or adopt a different arrangement of components.

[0235] In some embodiments, a computer-readable storage medium is further provided, wherein a computer program is stored in the storage medium, and when the computer program is executed by a processor, the steps of the fingerprint biometric detection method of the above-described embodiment are implemented. For example, the computer-readable storage medium may be a ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc.

[0236] It should be noted that the computer-readable storage medium mentioned in the embodiments of the present application may be a non-volatile storage medium, i.e., a non-transient storage medium.

[0237] It should be understood that all or part of the steps for implementing the above embodiments may be implemented by software, hardware, firmware, or any combination thereof. When implemented by software, all or part may be implemented in the form of a computer program product. A computer program product includes one or more computer instructions. The computer instructions may be stored on the aforementioned computer-readable storage medium.

[0238] That is, in some embodiments, a computer program product including a command is further provided, so that when this is executed on a computer, the computer performs the steps of the fingerprint biometric detection method described above.

[0239] The embodiments of this application mentioned above are not intended to limit this application, and any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of this application are all included within the scope of protection of this application.

Claims

Claim 1 A fingerprint biometric detection method performed by a fingerprint biometric detection device, comprising: a step of acquiring a first fingerprint image and a second fingerprint image - wherein the first fingerprint image and the second fingerprint image are images of the same fingerprint, the first fingerprint image is a red channel image and the second fingerprint image is a blue channel image, and the image sizes of the first fingerprint image and the second fingerprint image are the same -; a step of performing region segmentation for each of the first fingerprint image and the second fingerprint image - wherein the segmented first fingerprint image includes a plurality of first predicted bright areas and a plurality of first predicted dark areas, and the segmented second fingerprint image includes a second predicted bright area and a plurality of second predicted dark areas -; a step of determining the number of pixel points corresponding to each grayscale value within the plurality of first predicted bright areas and the plurality of second predicted dark areas; a step of generating a first grayscale feature vector based on the number of pixel points corresponding to each grayscale value within the plurality of first predicted bright areas and the plurality of second predicted dark areas; and a step of determining the number of pixel points corresponding to each grayscale value within the plurality of first predicted dark areas and the plurality of second predicted bright areas. A fingerprint biometric detection method characterized by comprising: a step of generating a second grayscale feature vector based on the number of pixel points corresponding to each grayscale value within a plurality of first predicted dark regions and a plurality of second predicted bright regions; a step of generating a multi-partition grayscale distribution feature vector based on the first grayscale feature vector and the second grayscale feature vector; and a step of inputting the multi-partition grayscale distribution feature vector into a fingerprint biometric detection model to obtain a fingerprint biometric detection result - wherein the fingerprint biometric detection model is obtained by training through supervised learning based on a biometric fingerprint sample image and a non-biometric fingerprint sample image. Claim 2 A fingerprint biometric detection method according to claim 1, wherein the step of acquiring a first fingerprint image and a second fingerprint image comprises: a step of acquiring a first original image - said first original image is an image of a fingerprint collected when the green light is turned off and the red light and blue light are turned on -; a step of generating the first fingerprint image by extracting image data of the red channel of the first original image; and a step of generating the second fingerprint image by extracting image data of the blue channel of the first original image; or a step of acquiring a second original image and a third original image - said second original image is a fingerprint image collected when the green light and blue light are turned off and the red light is turned on, and the third original image is a fingerprint image collected when the green light and red light are turned off and the blue light is turned on -; a step of generating the first fingerprint image by extracting image data of the red channel of the second original image; and a step of generating the second fingerprint image by extracting image data of the blue channel of the third original image. Claim 3 A fingerprint biometric detection method according to claim 1, wherein the step of performing region segmentation for each of the first fingerprint image and the second fingerprint image comprises: a step of determining partition parameters based on the distribution rules of bright and dark regions of the red channel image and the blue channel image of the biometric fingerprint, wherein the partition parameters include a plurality of partition width ratios; and a step of performing region segmentation for each of the first fingerprint image and the second fingerprint image based on the plurality of partition width ratios. Claim 4 A fingerprint biometric detection method according to claim 1, further comprising, prior to the step of generating the first grayscale feature vector based on the number of pixel points corresponding to each grayscale value within the plurality of first predicted bright areas and the plurality of second predicted dark areas, a step of obtaining a grayscale value statistical range - wherein the grayscale value statistical range is determined based on exposure parameters of an image acquisition device that collects the collected image of the fingerprint -; and the step of generating the first grayscale feature vector based on the number of pixel points corresponding to each grayscale value within the plurality of first predicted bright areas and the plurality of second predicted dark areas, a step of determining, based on the grayscale value statistical range, the number of pixel points corresponding to each grayscale value within the grayscale value statistical range within the plurality of first predicted bright areas and the plurality of second predicted dark areas; and a step of generating the first grayscale feature vector based on the number of pixel points corresponding to each grayscale value within the grayscale value statistical range within the plurality of first predicted bright areas and the plurality of second predicted dark areas. Claim 5 A fingerprint biometric detection method according to claim 1, further comprising, prior to the step of inputting the multi-partition grayscale distribution feature vector into a fingerprint biometric detection model to obtain a fingerprint biometric detection result, the step of acquiring the biometric fingerprint sample image and the non-biometric fingerprint sample image - wherein the biometric fingerprint sample image includes images of a plurality of biometric fingerprint samples collected with the green light turned off and the red light and blue light turned on, and the non-biometric fingerprint sample image includes images of a plurality of non-biometric fingerprint samples collected with the green light turned off and the red light and blue light turned on -; and the step of training the fingerprint biometric detection model through supervised learning based on the biometric fingerprint sample image and the non-biometric fingerprint sample image. Claim 6 A fingerprint biometric detection method according to claim 1, further comprising: a step of acquiring a fourth original image - said fourth original image is an image of the fingerprint collected while a green light is turned on -; a step of determining a fingerprint recognition result based on said fourth original image; and a step of determining a security verification result based on said fingerprint recognition result and said fingerprint biometric detection result. Claim 7 As a fingerprint biometric detection device, the fingerprint biometric detection device comprises an image collector and a processor, wherein the image collector comprises a fill light and an image sensor; the image collector collects a first original image; the fill light fills the image sensor with light by turning off a green light and turning on a red light and a blue light during the process of collecting the first original image; the processor processes the first original image to acquire a first fingerprint image and a second fingerprint image, wherein the first fingerprint image is a red channel image and the second fingerprint image is a blue channel image, and the image sizes of the first fingerprint image and the second fingerprint image are the same; region segmentation is performed for each of the first fingerprint image and the second fingerprint image, wherein the segmented first fingerprint image includes a plurality of first predicted bright areas and a plurality of first predicted dark areas, and the segmented second fingerprint image includes a second predicted bright area and a plurality of second predicted dark areas; and the number of pixel points corresponding to each grayscale value within the plurality of first predicted bright areas and the plurality of second predicted dark areas is determined; Generating a first grayscale feature vector based on the number of pixel points corresponding to each grayscale value within the plurality of first predicted bright areas and the plurality of second predicted dark areas; determining the number of pixel points corresponding to each grayscale value within the plurality of first predicted dark areas and the plurality of second predicted bright areas; and generating a second grayscale feature vector based on the number of pixel points corresponding to each grayscale value within the plurality of first predicted dark areas and the plurality of second predicted bright areas.A fingerprint biometric detection device characterized by generating a multi-partition grayscale distribution feature vector based on the first grayscale feature vector and the second grayscale feature vector, inputting the multi-partition grayscale distribution feature vector into a fingerprint biometric detection model to obtain a fingerprint biometric detection result, wherein the fingerprint biometric detection model is obtained by training through supervised learning based on a biometric fingerprint sample image and a non-biometric fingerprint sample image. Claim 8 A fingerprint biometric detection device according to claim 7, wherein the image collector is further used to collect a fourth original image; the fill light is further used to turn on a green light during the process of collecting the fourth original image to fill the image sensor with light; and the processor is further used to determine a fingerprint recognition result based on the fourth original image and to determine a security verification result based on the fingerprint recognition result and the fingerprint biometric detection result. Claim 9 A computer-readable storage medium, wherein a computer program is stored in the storage medium, and when the computer program is executed by a processor, a step of the method of any one of claims 1 to 6 is implemented. Claim 10 delete Claim 11 delete Claim 12 delete Claim 13 delete Claim 14 delete

Images